/*Some manual manipulation of tables is needed to produce the published versions*/

/*whole sample*/
eststo clear
/* belief variables: Likert 1-5, lower number means agree more that waste causes problem
*/
reg ch_nature video text, nocon
test video=text
eststo naturechg,title ("Reducing waste is a major way to conserve natural resources") add(p_value r(p))
reg ch_nature video 
reg ch_litter video text , nocon
test video=text
eststo litterchg,title ("Reducing waste is a major way to reduce litter") add(p_value r(p))
reg ch_litter video 
reg ch_energy video text, nocon
test video=text
eststo energychg,title ("Reducing waste is a major way to conserve energy") add(p_value r(p))
reg ch_energy video 
reg ch_landfills video text, nocon
test video=text
eststo landfillschg,title ("Reducing waste is a major way to reduce landfills") add(p_value r(p))
reg ch_landfills video
esttab naturechg litterchg energychg landfillschg using TableY2, b(2) scalars(p_value) not star nonum noobs label title("Table Y2: Changes in beliefs for those without strong pro-environmental responses before treatment") addnotes("for tests of zero means") mtitles rtf replace

/*attitudes, sense of responsibility, and personal norms
good: continuous scale 1-5, higher means reducing, reusing, recycling is more good
obligation: continuous scale 1-5, higher means stronger obligation to reduce/reuse/recycle
guilty: continuous scale 1-5, lower means stronger feeling of guilt if they don't reduce/reuse/recycle
*/
reg ch_good video text,nocon
test video=text
eststo goodchg, title("Reducing waste is good") add(p_value r(p))
reg ch_good video 
reg ch_obligation video text,nocon
test video=text
eststo obligchg, title ("Strong obligation to reduce waste") add(p_value r(p))
reg ch_obligation video 
reg ch_guilty video text, nocon
test video=text
eststo guiltychg, title ("Stronger feeling of guilt") add(p_value r(p))
reg ch_guilty video 
esttab goodchg obligchg guiltychg using TableY3, b(2) scalars(p_value) not star nonum noobs label title("Table Y3: Changes in attitudes, sense of responsibility, and personal norms for those without strong pro-environmental responses before treatment") addnotes("for tests of zero means") mtitles rtf replace

/*intended variables: Likert 1-5, higher number means more likely to reduce/reuse/recycle
*/
reg ch_intended video text, nocon
test video=text
eststo intchg, title ("More reduce/reuse/recycle") add(p_value r(p) ,replace)
reg ch_intended video
esttab intchg using TableY4, b(2) scalars(p_value) not star nonum noobs label title("Table Y4: Changes in intended behavior") addnotes("for tests of zero means") mtitles rtf replace


/*restricted sample*/
eststo clear

/* belief variables: Likert 1-5, lower number means agree more that waste causes problem
*/
reg ch_nature video text if nature_pre >2, nocon
test video=text
eststo naturechg,title ("Reducing waste is a major way to conserve natural resources") add(p_value r(p))
reg ch_nature video if nature_pre >2
reg ch_litter video text if litter_pre >2, nocon
test video=text
eststo litterchg,title ("Reducing waste is a major way to reduce litter") add(p_value r(p))
reg ch_litter video if litter_pre >2
reg ch_energy video text if energy_pre >2, nocon
test video=text
eststo energychg,title ("Reducing waste is a major way to conserve energy") add(p_value r(p))
reg ch_energy video if energy_pre >2
reg ch_landfills video text if landfills_pre >2, nocon
test video=text
eststo landfillschg,title ("Reducing waste is a major way to reduce landfills") add(p_value r(p))
reg ch_landfills video if landfills_pre >2
esttab naturechg litterchg energychg landfillschg using TableX2, b(2) scalars(p_value) not star nonum label title("Table X2: Changes in beliefs for those without strongly pro-environmental responses before treatment") addnotes("for tests of zero means") mtitles rtf replace

/*attitudes, sense of responsibility, and personal norms
good: continuous scale 1-5, higher means reducing, reusing, recycling is more good
obligation: continuous scale 1-5, higher means stronger obligation to reduce/reuse/recycle
guilty: continuous scale 1-5, lower means stronger feeling of guilt if they don't reduce/reuse/recycle
*/
reg ch_good video text if good_pre < 3.5,nocon
test video=text
eststo goodchg, title("Reducing waste is good") add(p_value r(p))
reg ch_good video if good_pre < 3.5
reg ch_obligation video text if obligation_pre < 3.5,nocon
test video=text
eststo obligchg, title ("Strong obligation to reduce waste") add(p_value r(p))
reg ch_obligation video if obligation_pre <3.5
reg ch_guilty video text if guilty_pre > 2.5, nocon
test video=text
eststo guiltychg, title ("Stronger feeling of guilt") add(p_value r(p))
reg ch_guilty video if guilty_pre > 2.5
esttab goodchg obligchg guiltychg using TableX3, b(2) scalars(p_value) not star nonum label title("Table X3: Changes in attitudes, sense of responsibility, and personal norms for those without strongly pro-environmental responses before treatment") addnotes("for tests of zero means") mtitles rtf replace

/*intended variables: Likert 1-5, higher number means more likely to reduce/reuse/recycle
*/
reg ch_intended video text if intended_pre <3.5, nocon
test video=text
eststo intchg, title ("More reduce/reuse/recycle") add(p_value r(p) ,replace)
reg ch_intended video if intended_pre < 3.5
esttab intchg using TableX4, b(2) scalars(p_value) not star nonum label title("Table X4: Changes in intended behavior for those without strongly pro-environmental responses before treatment") addnotes("for tests of zero means") mtitles rtf replace



